There are two types of data — Quantitative and Qualitative. No, there are four types of data – Nominal, Ordinal, Interval and Ratio. And this confusion does not stop here, as the question of the types of data is subjective. The key to effectively answering this question is defining the basis for dividing the data types. We will now look at all the possible basis on which the data can be categorized and discuss them individually.
Based on the characteristic:
1. Quantitative Data: This data type is related to quantity, as the name suggests. Data are said to be Quantitative Data if a numerical quantity is associated with each observation, for, e.g. Weight of students in a university. Generally, interval or ratio scales are used as a measurement of scale in the case of Quantitative Data.
2. Qualitative Data: This data type relates to the quality of an object or thing being studied. Because quality cannot be quantified numerically, it can only be calculated based on its presence or absence. If the characteristic under consideration is ‘color,’ then the objects can be classified as Red, Indigo, Blue, and so on. In the case of Qualitative Data, nominal and ordinal scales are commonly used as a measurement of scale.
Based on the nature of characteristic:
1. Discrete Data: If the nature of the characteristic under study is such that the values of observations can only be counted between two specific limits, then the corresponding data are referred to as Discrete Data. E.g. the number of books in a library, the number of children in a family, and so on.
2. Continuous Data: Continuous Data are those in which the measurement of observations of a characteristic under study can be any real value between two specific limits. E.g. Data obtained by measuring the heights or weights of the students of a particular class.
Based on the level of measurement:
1. Nominal Data: Under a nominal scale, we classify or divide the objects under study into two or more categories by giving them unique names. E.g. Classification into different categories based on gender, such as Male, Female or classification into different categories based on caste, such as General, OBC, ST etc.
Note: It is important to note that even if we code each category as a numerical value, the order does not matter. E.g. if Male is coded as ‘1’ and Female as ‘0’, then 1>0 or Male>Female does not matter.
2. Ordinal Data: As seen in the note, the order does not have any meaning in Nominal Data. As the name suggests, Ordinal Data addresses this. Other than the names or codes given to the different categories, Ordinal Data also provides the order among them. Using the ordinal scale, we can place the objects in a series based on the orders or ranks. But here, we cannot find the actual difference between the two categories. E.g. Suppose, a schoolboy is asked to list the name of three ice cream flavors according to his preference. Suppose he lists them in the following order: Vanilla > Straw berry > Tooty-fruity. This indicates that he likes vanilla more than strawberry and strawberry more than tooty-frooty. But we cannot measure the actual difference between his liking of vanilla and strawberry.
3. Interval Data: Interval Data is a type of data which is measured along a scale, in which each point is placed at an equal distance (interval) from one another. An example of interval data is the data collected about student test scores for grading. Different ranges of scores can be graded as distinction, first grade, second grade etc.
4. Ratio Data: This data type includes all the properties of the last three data types with one more property by providing a natural zero (absolute zero). E.g. suppose there are 60 teachers in a particular school in Mumbai. If we associate a unique number to each teacher related to the cash (in rupees), the teacher has with them at the time of the investigation. Then we have a fixed whole number corresponding to each teacher. Of course, two or more teachers may have the same cash (in rupees), but the teachers who have the same whole number will fall into one category. Here we note that the whole numbers allotted to the teachers can be ordered, have an actual difference and also have an origin (i.e. absolute zero ‘0’). Here natural zero indicates the absence of money in the teacher’s pocket.
Based on the time component:
1. Time Series Data: If the purpose of data collection has its connection with time then it is known as Time Series Data. Daily profit over a period of five years.
2. Cross-sectional Data: We are sometimes curious about how a characteristic under study at one point in time (such as income or expenditure, population, votes in an election, etc.) is distributed across different subjects (such as families, countries, political parties, etc.). Cross-sectional Data is data that is collected at a single point in time. For example, the annual income of various families in a community, consumer expenditure survey conducted by a research scholar, opinion polls conducted by an agency, salaries of all employees of an institute, and so on.
Based on ways of obtaining data:
1. Primary Data: Data collected by an investigator, agency or institution for a specific purpose and these people are first to use these data, are called Primary Data. E.g. suppose a research scholar wants to know the mean age of students of M.Sc. Economics of a particular university.
2. Secondary Data: Data obtained by an investigator, agency, or institution from an existing source are called Secondary Data. Meaning this data was initially collected by someone and has been used by them at least once.
#DataTypes #DataAnalysis #DataExploration #DataScience #DataAnalytics #DataVisualization #BigData #StructuredData #UnstructuredData #DataClassification #Statistics #AppliedStatistics
References:
https://researchguides.ben.edu/c.php?g=282050&p=4036581#:~:text=Primary%20data%20refers%20to%20the,collected%20by%20someone%20else%20earlier.&text=Surveys%2C%20observations%2C%20experiments%2C%20questionnaire,journal%20articles%2C%20internal%20records%20etc
https://studyonline.unsw.edu.au/blog/types-of-data#:~:text=Psychologist%20Stanley%20Stevens%20developed%20the,%2C%20ordinal%2C%20interval%20and%20ratio